Project summary: While research shows that dysfunctional neighborhoods are often associated with multiple negative outcomes in adolescents (including substance use), not all individuals from such neighborhoods develop problem behaviors. Still, the specific mechanisms connecting dysfunctional neighborhoods with adolescent substance use, as well as with unexpected resilient outcomes have not been adequately explored in the literature. The proposed project aims to bridge this gap, by examining how adolescents' neighborhoods, families and their individual characteristics relate both positively and negatively to their substance use. We argue that through the process of social contagion, neighborhood dysfunction acts as an independent developmental risk, to directly promote anti-social problems among adolescents exposed to it. We also argue that there is a developmental progression of these anti-social problems, with adolescent substance use being preceded by school and behavioral problems. Finally, we argue that these negative associations between high-risk neighborhoods and adolescent outcomes can be alleviated by positive family relationships, which are a protective factor against neighborhood dysfunction. These hypotheses will be tested using the existing data from the Springfield Child Development Project (SCDP), a prospective longitudinal study of a community-representative sample of children and their mothers. The proposed R03 will analyze the final two SCDP assessments (T3 and T4; N - 357, in which adolescent substance use was assessed), and will examine how neighborhoods, families, and adolescents' own behavior impact their substance use over a one year follow up period. Using Structural Equation Modeling (SEM), we will first determine whether T3 neighborhood dysfunction is a risk for adolescent substance use one year later, after controlling for prior substance use and demographics. Next, we will determine whether adolescents' other anti-social problems at T3 mediate the risk effect of T3 neighborhood dysfunction on T4 substance use. Finally, we will determine whether positive family relationships have protective effects against the risk of dysfunctional neighborhoods, by separately analyzing the associations between T3 neighborhood dysfunction, T3 anti-social problems, and T4 substance use in adolescents who (a) do have, and (b) do not have positive family relationships. Relevance: If risk effects of neighborhood dysfunction on substance use are lessened in adolescents with positive family relationships (compared to adolescents without such positive relationships), then support will be obtained for the hypothesis that positive family relationships are a protective factor against neighborhood dysfunction, buffering its risk effects and reducing the serious public health problem of adolescent substance use. Consequently, more informed preventive efforts can be developed and aimed at families from high-risk neighborhoods and at children who are at risk for, but have not yet engaged in substance use. [unreadable] [unreadable]